a variant of the large sieve inequality with explicit
play

A variant of the large sieve inequality with explicit constants - PowerPoint PPT Presentation

A variant of the large sieve inequality with explicit constants Maciej Grzekowiak Adam Mickiewicz University Pozna, Poland Number Theoretic Methods in Cryptology Paris 2019 MG (UAM Pozna) Sieve NutMic 2019 1 / 25 Outline 1 The large


  1. A variant of the large sieve inequality with explicit constants Maciej Grześkowiak Adam Mickiewicz University Poznań, Poland Number Theoretic Methods in Cryptology Paris 2019 MG (UAM Poznań) Sieve NutMic 2019 1 / 25

  2. Outline 1 The large sieve inequality MG (UAM Poznań) Sieve NutMic 2019 2 / 25

  3. Outline 1 The large sieve inequality 2 The algorithmic number theory problem MG (UAM Poznań) Sieve NutMic 2019 2 / 25

  4. Outline 1 The large sieve inequality 2 The algorithmic number theory problem 3 Application of the large sieve inequality MG (UAM Poznań) Sieve NutMic 2019 2 / 25

  5. Outline 1 The large sieve inequality 2 The algorithmic number theory problem 3 Application of the large sieve inequality MG (UAM Poznań) Sieve NutMic 2019 2 / 25

  6. The large sieve inequality We define M + N � e ( θ ) = e 2 π i θ , S ( x ) = c n e ( nx ) , n = M + 1 where the c n are arbitrary complex numbers. MG (UAM Poznań) Sieve NutMic 2019 3 / 25

  7. The large sieve inequality We define M + N � e ( θ ) = e 2 π i θ , S ( x ) = c n e ( nx ) , n = M + 1 where the c n are arbitrary complex numbers. The distance to nearest integer function � θ � = min {| θ − n | : n ∈ Z } MG (UAM Poznań) Sieve NutMic 2019 3 / 25

  8. The large sieve inequality Let x 1 , . . . x R be points which are well spaced modulo 1 in the sense that � x r − x s � ≥ δ (1) for s � = r , where 0 < δ ≤ 1 2 . MG (UAM Poznań) Sieve NutMic 2019 4 / 25

  9. The large sieve inequality Let x 1 , . . . x R be points which are well spaced modulo 1 in the sense that � x r − x s � ≥ δ (1) for s � = r , where 0 < δ ≤ 1 2 . The large sieve is an inequality of the form R M + N | S ( x r ) | 2 ≤ ∆ � � | c n | 2 , (2) r = 1 n = M + 1 where ∆ = ∆( N , δ ) . MG (UAM Poznań) Sieve NutMic 2019 4 / 25

  10. The large sieve inequality Let x 1 , . . . x R be points which are well spaced modulo 1 in the sense that � x r − x s � ≥ δ (1) for s � = r , where 0 < δ ≤ 1 2 . The large sieve is an inequality of the form R M + N | S ( x r ) | 2 ≤ ∆ � � | c n | 2 , (2) r = 1 n = M + 1 where ∆ = ∆( N , δ ) . [Gallagher] For example, we can take ‘ ∆ = π N + δ − 1 MG (UAM Poznań) Sieve NutMic 2019 4 / 25

  11. Application of the large sieve inequality Let x r = a q be points, where ( a , q ) = 1 , q ≤ Q , . If a q � = a ′ q ′ then MG (UAM Poznań) Sieve NutMic 2019 5 / 25

  12. Application of the large sieve inequality Let x r = a q be points, where ( a , q ) = 1 , q ≤ Q , . If a q � = a ′ q ′ then aq ′ − a ′ q q − a ′ � � � � a qq ′ ≥ 1 1 � � � � � = � ≥ � � � � q ′ qq ′ Q 2 � � We may take δ = Q − 2 , we obtain MG (UAM Poznań) Sieve NutMic 2019 5 / 25

  13. Application of the large sieve inequality Lemma q M + N | S ( a / q ) | 2 ≤ ( N + Q 2 ) � � � | c n | 2 , q ≤ Q a = 1 n = M + 1 ( a , q )= 1 where the summation is over primes q . MG (UAM Poznań) Sieve NutMic 2019 6 / 25

  14. Application of the large sieve inequality Let π ( x ; a , q ) = ♯ { p ≤ x : p ≡ a (mod q ) , ( a , q ) = 1 } MG (UAM Poznań) Sieve NutMic 2019 7 / 25

  15. Application of the large sieve inequality Let π ( x ; a , q ) = ♯ { p ≤ x : p ≡ a (mod q ) , ( a , q ) = 1 } Then 2 y � � log log( 3 y / q ) �� π ( x + y ; a , q ) − π ( x ; a , q ) ≤ 1 + O ϕ ( q ) log( y / q ) log( 2 y / q ) for y > q . MG (UAM Poznań) Sieve NutMic 2019 7 / 25

  16. Application of the large sieve inequality Let M + N � T ( χ ) = c n χ ( n ) n = M + 1 where χ is a Dirichlet character (mod q ) . MG (UAM Poznań) Sieve NutMic 2019 8 / 25

  17. Application of the large sieve inequality Let M + N � T ( χ ) = c n χ ( n ) n = M + 1 where χ is a Dirichlet character (mod q ) . Gallagher show q | T ( χ ) | 2 ≤ ϕ ( q ) � ∗ � | S ( a / q ) | 2 q χ mod q a = 1 ( a , q )= 1 where � ∗ denotes summation over primitive multiplicative characters χ (mod q ) . MG (UAM Poznań) Sieve NutMic 2019 8 / 25

  18. Application of the large sieve inequality We obtain M + N q � ∗ | T ( χ ) | 2 ≤ ( N + Q 2 ) � � | c n | 2 , ϕ ( q ) χ mod q n = M + 1 q ≤ Q where the summation is over primes q and � ∗ denotes summation over primitive multiplicative characters χ (mod q ) . MG (UAM Poznań) Sieve NutMic 2019 9 / 25

  19. Generalization of the large sieve inequality Huxley generalized to algebraic number fields K , [ K : Q ] = k . MG (UAM Poznań) Sieve NutMic 2019 10 / 25

  20. Generalization of the large sieve inequality Huxley generalized to algebraic number fields K , [ K : Q ] = k . He considered algebraic integers of α ∈ K such that α = n 1 ω 1 + . . . + n k ω k , M i + 1 ≤ n i ≤ M i + N i , i = 1 , . . . k , MG (UAM Poznań) Sieve NutMic 2019 10 / 25

  21. Generalization of the large sieve inequality Huxley generalized to algebraic number fields K , [ K : Q ] = k . He considered algebraic integers of α ∈ K such that α = n 1 ω 1 + . . . + n k ω k , M i + 1 ≤ n i ≤ M i + N i , i = 1 , . . . k , Schaal considered α ∈ K lying in the domains which not necessarily depend on special integer basis of K . MG (UAM Poznań) Sieve NutMic 2019 10 / 25

  22. Generalization of the large sieve inequality Huxley generalized to algebraic number fields K , [ K : Q ] = k . He considered algebraic integers of α ∈ K such that α = n 1 ω 1 + . . . + n k ω k , M i + 1 ≤ n i ≤ M i + N i , i = 1 , . . . k , Schaal considered α ∈ K lying in the domains which not necessarily depend on special integer basis of K . Hinz proved a variant of the large sieve inequality to algebraic number K MG (UAM Poznań) Sieve NutMic 2019 10 / 25

  23. Problem Find two primes p and q such that q | ♯ E ( F p ) . MG (UAM Poznań) Sieve NutMic 2019 11 / 25

  24. Problem Find two primes p and q such that q | ♯ E ( F p ) . Our assumptions p should be as close to q as possible MG (UAM Poznań) Sieve NutMic 2019 11 / 25

  25. Problem Find two primes p and q such that q | ♯ E ( F p ) . Our assumptions p should be as close to q as possible works in a polynomial time with respect to p , MG (UAM Poznań) Sieve NutMic 2019 11 / 25

  26. Problem Find two primes p and q such that q | ♯ E ( F p ) . Our assumptions p should be as close to q as possible works in a polynomial time with respect to p , give a proof without assumptions of any hypotheses, any heuristics, MG (UAM Poznań) Sieve NutMic 2019 11 / 25

  27. Problem Find two primes p and q such that q | ♯ E ( F p ) . Our assumptions p should be as close to q as possible works in a polynomial time with respect to p , give a proof without assumptions of any hypotheses, any heuristics, compute the order of magnitude of p , q for which we can proof that the algorithm works MG (UAM Poznań) Sieve NutMic 2019 11 / 25

  28. Application. Elliptic Curve Cryptography (ECC) Theorem [Shparlinski, Sutherland 2014] Given a real number x > 3.There is an Algorithm that outputs p ∈ [ x , 2 x ] , a , b ∈ F p , N = ♯ E ( F p ) , where p is uniformly distributed over primes in [ x , 2 x ] and the pair ( a , b ) is then uniformly distributed over pairs in F p × F p for which ♯ E ( F p ) is prime. Assuming the GRH, the expected running time of the Algorithm is O ((log x ) 5 (log log x ) 3 log log log x ) MG (UAM Poznań) Sieve NutMic 2019 12 / 25

  29. Application. Elliptic Curve Cryptography (ECC) Theorem [Shparlinski, Sutherland 2017] Assume the GRH. There is a deterministic algorithm that, given a prime p and an integer m = o ( p 1 / 2 (log p ) − 4 ) , outputs an elliptic curve E ( F p ) with m | ♯ E ( F p ) in O ( mp 1 / 2 ) time. MG (UAM Poznań) Sieve NutMic 2019 13 / 25

  30. Application. Elliptic Curve Cryptography (ECC) CM method: MG (UAM Poznań) Sieve NutMic 2019 14 / 25

  31. Application. Elliptic Curve Cryptography (ECC) CM method: select p , MG (UAM Poznań) Sieve NutMic 2019 14 / 25

  32. Application. Elliptic Curve Cryptography (ECC) CM method: select p , find ∆ < 0 and s , t ∈ Z such that 4 p = t 2 − ∆ s 2 , MG (UAM Poznań) Sieve NutMic 2019 14 / 25

  33. Application. Elliptic Curve Cryptography (ECC) CM method: select p , find ∆ < 0 and s , t ∈ Z such that 4 p = t 2 − ∆ s 2 , If p + 1 ± t is a prime, then construct E , or MG (UAM Poznań) Sieve NutMic 2019 14 / 25

  34. Application. Elliptic Curve Cryptography (ECC) CM method: select p , find ∆ < 0 and s , t ∈ Z such that 4 p = t 2 − ∆ s 2 , If p + 1 ± t is a prime, then construct E , or If p + 1 ± t has a big prime factor q , then construct E , MG (UAM Poznań) Sieve NutMic 2019 14 / 25

  35. CM-primes MG (UAM Poznań) Sieve NutMic 2019 15 / 25

  36. CM-primes DEFINITION: MG (UAM Poznań) Sieve NutMic 2019 15 / 25

  37. CM-primes DEFINITION: Primes p and q are CM-primes with respect to ∆ < 0 if MG (UAM Poznań) Sieve NutMic 2019 15 / 25

  38. CM-primes DEFINITION: Primes p and q are CM-primes with respect to ∆ < 0 if there exist integers s and t such that | t | ≤ 2 √ p , 4 p − t 2 = ∆ s 2 . q | p + 1 − t , MG (UAM Poznań) Sieve NutMic 2019 15 / 25

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend


More recommend